I Built an Office Full of AI Developers. They Started Opening Pull Requests.

Описание и применение

Разработчик создал «Мой маленький офис», где ИИ‑агенты автономно работают с реальным репозиторием, но только после прохождения тестов открывают пулл‑реквесты, тем самым меняя роль разработчика с писателя к руководителю задач.

Задача

Developers need to manage AI agents that autonomously modify code, ensuring correctness, safety, and coordination.

Результат

Agents can autonomously modify code, but verification gates and isolated workspaces prevent destructive changes, ensuring only passing tests lead to PRs.

Рабочий процесс

• Define a task for an agent (e.g., "Fix checkout crash").
• Agent clones repository into isolated workspace.
• Agent understands task, inspects code, makes changes, runs project checks.
• If checks fail, agent attempts repair; if still failing, task is marked failed.
• If checks pass, agent commits changes, pushes branch, and opens pull request.
• Human reviews PR and verifies final outcome.
• Deploy to production if approved.

Доказательства

"If the project's verification commands fail, the task doesn't get delivered."

"An agent can work inside its workspace. It cannot casually wander around your machine."

"Tests become the fucking contract."